"""FTS5 search engine for snippets."""

from typing import Any

from snip.db.database import Database, get_database


class SearchEngine:
    def __init__(self, db: Database | str | None = None):
        if isinstance(db, str) or db is None:
            self.db = get_database(db)
        else:
            self.db = db

    def search(
        self,
        query: str,
        limit: int = 50,
        language: str | None = None,
        tag: str | None = None,
    ) -> list[dict[str, Any]]:
        """Search snippets using FTS5."""
        return self.db.search_snippets(query, limit=limit, language=language, tag=tag)

    def highlight(self, text: str, query: str) -> str:
        """Add highlighting markers around matched terms."""
        import re
        terms = re.split(r'\s+', query)
        result = text
        for term in terms:
            if term:
                result = re.sub(f'({re.escape(term)})', r'**\1**', result, flags=re.IGNORECASE)
        return result

    def suggest(self, prefix: str, limit: int = 10) -> list[str]:
        """Suggest completions for a prefix."""
        snippets = self.db.list_snippets(limit=100)
        suggestions = set()
        for s in snippets:
            title = s.get("title", "")
            if title.lower().startswith(prefix.lower()):
                suggestions.add(title)
            tags = s.get("tags", [])
            if isinstance(tags, str):
                tags = json.loads(tags)
            for tag in tags:
                if tag.lower().startswith(prefix.lower()):
                    suggestions.add(tag)
        return sorted(list(suggestions))[:limit]

    def parse_query(self, query: str) -> dict[str, Any]:
        """Parse a search query into components."""
        import re
        result = {
            "terms": [],
            "language": None,
            "tag": None,
        }
        language_match = re.search(r'language:(\w+)', query)
        if language_match:
            result["language"] = language_match.group(1)
            query = re.sub(r'language:\w+', '', query)
        tag_match = re.search(r'tag:(\w+)', query)
        if tag_match:
            result["tag"] = tag_match.group(1)
            query = re.sub(r'tag:\w+', '', query)
        result["terms"] = query.split()
        return result